A new update to the company’s privacy policy says its AI can scrape any web content to improve its tools.
A new Google privacy policy update says that it will be allowed to scrape the web for any content it feels will be beneficial to enhancing and building its artificial intelligence tools.
The company will be using any part of the web it wants to teach and train its AI technology.
“Google uses information to improve our services and to develop new products, features, and technologies that benefit our users and the public,” states the update to the Google privacy policy. “For example, we use publicly available information to help train Google’s AI models and build products and features like Google Translate, Bard, and Cloud AI capabilities.”
According to a Gizmodo report, the update to the privacy policy was made to say: “AI models”, where the wording had previously stated “for language models.” Furthermore, the update also added the terms Bard and Cloud AI to its features list, which had previously mentioned only Google Translate, for which it stated that it collected data.
The update indicates that anything anyone posts online is subject to use by Google’s artificial intelligence.
The updated privacy policy indicates that if anyone creates and posts anything online, Google has the right to use it to train its AI models. The wording of the policy also indicates that it is not only content posted within the Google ecosystem in some way, but that the technology can access any information posted on any part of the web.
The Gizmodo report also pointed out that Google can also apply the new wording of the policy to allow it to collect old content that was still human-generated, such as old blog posts or reviews, in order to teach its AI about human speech and text’s development and distribution.
It should be noted that precisely how Google will use the data it collects has yet to be revealed. A number of social media platforms, such as Reddit and Twitter – considerable sources of current information – have already set limitations to their public access due to the massive sudden popularity of artificial intelligence chatbots.